Active Illumination Of The Target Image Simulation And Extraction Algorithm

It is difficult to track air targets in complex background. Active laser illumination is an important method to improve the stability of target extraction and the accuracy of character calculation.Some facts,such as illumination intensity,fluctuating of background and noise,let it worst in detecting target with active illumination. To improve the stability of target tracking,we had worked on the follows:(1) Computer simulation technology in active illumination environment. (2) Research of target extract algorithm.Using Kalman model and gaussian random phase-screen FFT,we simulated the distribution of illumination beam through long distance atmosphere and its frequency character. Also other facts in simulation were analysised,such as reflection of niaterial under natural lighting,point spread function of optic system and electronic noise. With OpenGL and VC++ language we made software that integrated all these facts above and compared it with experiment result to prove it's validation.Also the method of noise estimation was discussed,and was used to analysis the character of illumination through atmosphere. Then the Markov Random Fields (MRF) theory was used to segment target df active illumination by light and Sobel-transform information in this article. To make MRF algorithm suitable in actual setting,some technologies were analysis to make extracting steady.According to the analysis of laser beam through atmosphere turbulence,we found that:1. Illumination laser's wave is stronger than that of natural lighting. 2. In strong turbulence (Cn2>lx10-15),the high frequency distribution like the distribution of white noise. 3. We can get better even illumination with multi-beams. 4. The lighting distribution and edge information of gaussian beam through atmosphere could be described by Weibull function,but under little noise condition we can describe the target area's edge character as Gaussian distribution.Through the analysis of extracting stabilization we found:It's suitable to use the lighting segment in little turbulence at night,butit's better to use the edge segment under bad environment such as strong turbulence,strong natural lighting and cloudy background.